Browse Articles

Impact of a real-time targeted expert outreach model on conversion of biomarker results to precision treatments.

e23390 Background: Previous publications reported low rates of converting positive biomarker results into appropriate treatments as a reason for precision oncology practice gaps. One key cause being awareness: physicians are inundated with the rapidly evolving treatment landscapes and struggle to recall clinical utility for rare or emerging biomarkers. Proactive engagement efforts to educate physicians on biomarker-dependent treatments can help alleviate this gap. Pathologists, key experts in biomarker testing/reporting, are ideally positioned to lead such initiatives. We therefore hypothesize that real-time, targeted outreach to physicians from pathologists following positive biomarker tests will help support timely and appropriate precision therapy delivery. The aim of this study is to evaluate the impact of this real-time expert outreach model in boosting awareness of the clinical utility of novel or rare biomarkers in solid tumors. Methods: Real-world data from unstructured testing reports were collected from > 500 laboratories across the US in 2025. Diaceutics proprietary Agentic Artificial Intelligence platform was applied to identify HIPAA-compliant positive cases for selected novel and rare actionable biomarkers (1-2% incidence rate in solid tumors). Physicians associated with these cases were identified as candidates for engagement by expert pathologists who performed timely review of case details ahead of the outreach and provided targeted awareness on the clinical utility of positive biomarker results. Results: Among 34,242 solid tumor patients who underwent biomarker testing in 2025, 67 anonymized patients potentially eligible for a novel precision therapy based on positive rare biomarker fusion, were found. Based on this data, physicians who ordered these tests were identified, and were later matched with a pathologist for an outreach awareness program. Connections between pathologists and physicians were successfully established for 37 outreaches (55%), whilst 14 (21%) are still ongoing and 16 were declared to be unsuccessful (24%). Among those successfully connected physicians, only 15% reported being fully aware of the biomarker’s clinical utility, whereas 85% had limited or no awareness. This lack of awareness poses a risk on many patients missing appropriate biomarker-directed treatments. Through the pathologists-led discussion, this expert outreach program enabled 23/37 patients (62%) to receive treatment aligned with their molecular profile, who might have otherwise be missed in precision oncology practice. Conclusions: Our preliminary data confirms this real-time targeted expert (pathologists) outreach after rare actionable biomarker detection was feasible and substantially supported fostering awareness on clinical utility of novel or rare biomarkers to physicians for delivering timely and appropriate precision therapy.

Enhancing Delivery of Tobacco Treatment in Oncology Care: Moving the Needle

Detection of organophosphorus pesticide residues in breast cancer tissue: A translational integrated environmental exposure study.

e12580 Background: Geographic variation in cancer incidence reflects differences in exposure to environmental carcinogens. Pesticides, particularly organophosphorus compounds, are extensively used in agriculture and are known to function as endocrine disruptors, hormone mimics, and chronic inflammatory mediators. While epidemiologic associations between pesticide exposure and breast cancer risk have been reported, direct evidence of pesticide accumulation within human breast cancer tissue remains limited. To explore, we performed tissue-level detection of pesticide residues in breast cancer specimens and peritumoral adipose tissue, coupled with a root cause analysis through assessment of pesticide use practices, awareness, and regulatory checks among farmers. Methods: Breast cancer tissue specimens (n=30) were analyzed for pesticide residues using high-resolution mass spectrometry. Analytical accuracy and detection reliability were validated using appropriate positive controls. The presence, diversity, and relative distribution of pesticide residues were assessed separately in tumor tissue and peritumoral adipose tissue to evaluate preferential accumulation patterns within the tumor microenvironment. In parallel, structured interviews were conducted with 50 farmers to assess knowledge on pesticide type, quantity and frequency of use, awareness of potential carcinogenic effects, residue monitoring practices, and familiarity with regulatory guidelines governing pesticide application and food safety. Results: A total of 49 distinct pesticide residues were detected across breast cancer tissue samples. Peritumoral adipose tissue demonstrated a higher number and greater relative concentration of pesticide residues compared with tumor tissue, consistent with lipophilic nature of organophosphorus pesticides. Farmer interviews revealed limited awareness of the carcinogenic potential of commonly used pesticides, including organophosphorus compounds. Selection and quantity of pesticide use were primarily guided by peer practices, local convention, and perceived yield benefits rather than standardized safety guidelines. Routine monitoring of pesticide residue on agricultural produce was uncommon in domestic supply chains, whereas export-oriented products followed stringent residue control measures from cultivation through post-harvest processing and packaging. Conclusions: In addition to detected pesticide at tissue level, root cause analysis demonstrates limited awareness on safe pesticide usage at the grassroots level, underscoring the need for improved education, standardized residue monitoring, and strengthened regulatory enforcement to mitigate long-term cancer risk.

Multi-ancestry, trans-generational GWAS meta-analysis of gestational diabetes and glycaemic traits during pregnancy reveals limited evidence of pregnancy-specific genetic effects

Real-world toxicity outcomes with adjuvant pembrolizumab following nephrectomy for renal cell carcinoma.

e23447 Background: Adjuvant pembrolizumab improves overall and disease-free survival in high-risk renal cell carcinoma (RCC) patients; however, treatment-related toxicity remains a serious concern. We evaluated toxicity patterns and associated survival outcomes among patients receiving adjuvant pembrolizumab following nephrectomy in a large real-world cohort. Methods: We conducted a retrospective cohort study in Epic Cosmos, identifying the largest-to-date RCC cohort undergoing nephrectomy who received post-nephrectomy adjuvant pembrolizumab. The study period spanned from the FDA approval date (Nov 17, 2021) to Dec 20, 2025, ensuring at least one month of follow-up prior to the query date. Patients were excluded if they had metastatic systemic therapy exposure and/or metastatic diagnosis codes at any time before or within 90 days after nephrectomy. Overall survival (OS) was estimated using the Kaplan–Meier method. Logistic regression models with false discovery rate (FDR) correction were used to identify factors associated with toxicity (initiation of post-adjuvant high-dose systemic corticosteroids was used as a proxy for clinically significant immune-related adverse events). Results: The cohort included 2,656 patients with a median age of 64 years; 67.8% were male, and 86.5% were White. During follow-up, 400 patients (15.1%) initiated high-dose systemic corticosteroids, which was higher than reported in the KEYNOTE-564 trial (7.4%). Steroid use was associated with significantly worse OS compared with no steroid use (OR 1.99, 95% CI 1.21–3.27; p < 0.01). Laboratory factors significantly associated with toxicity requiring high-dose steroids included high eosinophils (OR: 2.44 [1.51-3.95], FDR < 0.01), eosinophil-to-lymphocyte ratio (OR: 2.44 [1.33-4.35], FDR = 0.02), RDW (OR: 1.07 [1.02-1.22], FDR = 0.01), and BUN (OR: 1.02 [1.003-1.03], FDR = 0.03). Higher values of albumin (OR: 0.62 [0.47-0.81], FDR < 0.01), hemoglobin (OR: 0.90 [0.84-0.96], FDR < 0.01), MCHC (OR: 0.89 [0.82-0.96], FDR = 0.02), and hematocrit (OR: 0.97 [0.94-0.99], FDR = 0.03) were associated with lower toxicity risk. None of the baseline clinical/ demographic features were significant, including age, sex, or BMI. Conclusions: In this large real-world cohort of patients with RCC receiving adjuvant pembrolizumab following nephrectomy, treatment-related toxicity requiring high-dose corticosteroids occurred in approximately 15.1% of patients and was associated with significantly worse overall survival. Baseline inflammatory and hematologic markers were associated with toxicity risk. These findings highlight the importance of early toxicity risk stratification and proactive management in the adjuvant immunotherapy setting.

From unstructured oncology EHRs to regulatory-grade real-world evidence: A next-generation registry approach.

e23425 Background: Real-world evidence (RWE) derived from oncology electronic health records (EHRs) is increasingly used to inform clinical research, healthcare decision-making, and regulatory discussions. However, most clinically relevant oncology information remains embedded in unstructured free text, limiting scalability, data quality, and reuse. Methods: We implemented the Savana Next-Generation Registry (SNGR), a multilingual software-as-a-service platform designed to generate clinical information from routine EHR data. Participating sites undergo standardized feasibility assessment, source-level pseudonymization, and predefined data quality checks. Heterogeneous EHR data are harmonized into a common data model and curated through automated multidimensional quality controls. Free-text narratives are structured using EHRead, a clinical natural language processing (cNLP) engine trained exclusively on real-world EHRs and integrating international medical terminologies enriched with real-world clinical variants. Terminology panels define key Population–Intervention–Outcome variables. A multilayer clinical quality assurance framework combining automated evaluation and human-in-the-loop validation ensures accuracy, reproducibility, and regulatory reliability. Results: SNGR has enabled over 65 real-world studies across multiple therapeutic areas, including oncology, generating more than 30 peer-reviewed publications. In oncology, SNGR has supported large multicenter registries involving millions of patients, enabling descriptive, comparative, and predictive analyses using routine clinical data. The platform has processed more than 550 million EHRs from over 33 million patients, supporting longitudinal, high-throughput data extraction. Embedded validation workflows consistently reported high precision, recall, and F1-scores for clinically relevant variables. Applications include characterization of treatment patterns and outcomes, prediction of thromboembolic and bleeding risk, and development of explainable prognostic models outperforming traditional staging systems. Conclusions: Next-generation registries based on validated cNLP enable scalable, reliable reuse of unstructured oncology EHR data to generate high-quality, regulatory-aligned RWE.

Modeling early gastric cancer evolution following <i>ARID1A</i> loss in Latino patient-derived organoids.

e16144 Background: Gastric cancer remains a leading cause of cancer mortality worldwide, with disproportionately worse outcomes among Latino patients in the United States. Despite this burden, Latino populations remain underrepresented in genomic datasets and preclinical models. ARID1A is one of the most frequently mutated tumor suppressor genes in gastric cancer, yet its role in early disease evolution remains incompletely understood. We sought to develop patient-derived gastric organoid models to understand how ARID1A loss contributes to early phenotypic progression. Methods: Normal gastric organoids were established from Latino patients and engineered using CRISPR/Cas9 to generate isogenic ARID1A/TP53 double-knockout (dKO) and TP53 knockout. Organoids were maintained in long-term culture for up to 350 days. Early (~150 days) and late (~350 days) time points were evaluated using histologic changes, RNA sequencing, and whole-exome sequencing to assess phenotypic progression and evolution. Functional assays evaluating proliferation, migration, and drug sensitivities are ongoing. Results: ARID1A/TP53 dKO organoids exhibited progressive morphological changes and increased proliferative features compared with wild-type controls. RNA sequencing demonstrated time-dependent changes, including suppression of digestion and iron transport pathways and upregulations associated with cilium movement and ciliogenesis. Whole exome sequencing revealed shared mutations across independently derived dKO clones, alongside acquisition of clone-specific variants over time. These alterations involved genes implicated in transcriptional regulation, extracellular matrix remodeling, and receptor signaling, consistent with early genomic instability and divergent clonal evolution. Conclusions: Patient-derived gastric organoids from Latino individuals demonstrate that combined ARID1A and TP53 loss drives transcriptional reprogramming and promotes early dysplastic progression. These models establish a translational platform to interrogate gastric cancer evolution and to identify therapeutic vulnerabilities associated with ARID1A loss. Collectively, these findings have implications for risk stratification, early detection, and precision prevention strategies in high-risk populations.

Correlation of tumor grade with increased CD3+ t-cell and FOXP3+ regulatory T-cell infiltration in neuroendocrine neoplasms.

e15186 Background: Neuroendocrine neoplasms (NENs) comprise a biologically heterogenous group of malignancies with diverse clinical behavior and outcomes. Tumor grade, defined by proliferative indices such as Ki-67, is a key determinant of prognosis and therapeutic decision-making across NEN subtypes. 1 Increasing evidence suggests that the tumor microenvironment (TME) influences tumor progression and treatment response. 2 Tumor-infiltrating CD3+cells mediate anti-tumor immunity, whereas FOXP3+ T-regulatory cells suppress effector T-cell activity and promote immune evasion. 3 While prior studies of the neuroendocrine tumor microenvironment have largely been limited to single organ systems, our study evaluates immune infiltration across a multi-site cohort of neuroendocrine neoplasms, enabling direct comparison of immune features by tumor grade. 4,5 We asked whether immune cell infiltration of high-grade NENs differs from low-grade NENs. Methods: We conducted a retrospective cohort study using archived, deidentified NEN specimens from Fox Chase Cancer Center. A tissue microarray comprising 19 low-grade (grades 1 and 2) and 19 high-grade (grade 3) NENs from multiple primary sites, including the gastrointestinal tract, pancreas, lung, bladder, ovary, and breast, was analyzed. Tumors were graded according to the World Health Organization criteria based on Ki-67 labeling index and mitotic rate. Immunohistochemical staining was performed for CD3 and FOXP3. Immune cell infiltration was quantified using AI-assisted digital image analysis (Visiopharm), with marker-positive cells normalized to the analyzed region of interest and expressed as a cell density (cells/mm 2 ). For cases with multiple tissue cores, density values were averaged to generate a single tumor-level value. Group comparisons were performed using two-sample Welch t-tests. All tests were two-sided, with p &lt; 0.05 considered statistically significant. Results: Immune cell infiltration differed significantly by tumor grade. G3 NENs demonstrated higher CD3+ T-cell density compared with G1/2 NENs (421.82 vs 132.22/mm 2 , p = 0.046). FOXP3+ T-reg density was also increased in G3 NENs (48.89 vs 9.95 cells/mm 2 , p = 0.029). Conclusions: High-grade neuroendocrine neoplasms exhibit increased infiltration of both CD3+ T-cells and FOXP3+ regulatory T-cells compared with low-grade tumors, suggesting a more immune-infiltrated yet potentially immunosuppressive tumor microenvironment with increasing tumor grade. These findings highlight tumor grade as correlate of immune microenvironment features in NENs and may have implications of immunotherapeutic strategies. Immune Cell Infiltration by Tumor Grade in Neuroendocrine Neoplasms. Mean CD3+cells/mm 2 Mean FOXP3+ cells/mm 2 High Grade 421.82 48.89 Low Grade 132.33 9.95 P-Value 0.046 0.029

E-PROs in oncology: Improving treatment adherence to treatment while reducing emergency care and costs for patients with gastrointestinal malignancies receiving systemic therapy.

11109 Background: Patients with cancer undergoing systemic therapy are prone to experience symptom-related complications which may lead to emergency room visits (ERV) and non-planned inpatient admissions (NIA), resulting in increased costs to health systems. Symptom monitoring and proactive management based on ePRO protocols have been shown to reduce acute care utilization; however, the economic impact of this intervention remains incompletely characterized. In addition, patients monitored through ePRO systems tend to remain on active anticancer treatment for longer periods, with fewer unplanned treatment interruptions. Our aim was to evaluate the economic impact of a remote Comprehensive Companion Program (CCP) in a private health system. Methods: We conducted a retrospective study within the AUNA oncological network in Lima, Peru. Adult patients with GI malignancies receiving systemic therapy who experienced at least one ERV or NIA between June 2021 and July 2022 were included. Patients were grouped according to enrollment in a remote CCP or receipt of usual care. Costs related to ERV and NIA were analyzed from a payer’s perspective and adjusted by time on treatment to estimate average monthly cost per-patient. Data were analyzed using descriptive and comparative statistical methods. Results: A total of 97 patients with comparable baseline clinical characteristics were included (50 in the CCP group and 47 with usual care). Patients in the CCP group had a significantly longer time on treatment (median 235.5 vs 151 days; p = 0.0009), driven primarily by patients with metastatic disease. When costs associated with ERV and NIA were normalized by treatment duration, the CCP group was associated with significantly lower monthly acute care costs. Median monthly variable costs per patient were USD $147.14 versus $449.70 (p = 0.0137), representing an 67% of costs reduction in monthly variable costs per patient. Similarly, median monthly prorated expenses per patient were $220.33 versus $636.66 (p = 0.0151), corresponding to an approximate 65% reduction in monthly prorated expenses. Conclusions: Our study showed that patients enrolled in a CCP incorporating e-PRO based symptom monitoring remained on treatment for longer; however, this was associated with significantly lower time-adjusted acute care costs. These findings highlight the potential value of the CCP in reducing cancer care–related costs, while acknowledging that the results are limited to a private health care setting.

Nivolumab plus lenvatinib for unresectable anaplastic thyroid cancer: Results of the phase 2 NAVIGATION study.

6021 Background: Anaplastic thyroid cancer (ATC) is one of the most aggressive and lethal malignancies, characterized by limited treatment options and a dismal prognosis. Although lenvatinib has shown some efficacy, its impact remains modest, and no standard systemic therapy has been established for unresectable ATC—particularly for BRAF wild-type (BRAF-negative) disease. Methods: The NAVIGATION study (NCT05696548) is an open-label, multicenter, phase 2 trial evaluating the combination of nivolumab and lenvatinib in patients with unresectable ATC. As primary endpoints, Step 1 assessed dose-limiting toxicities (DLTs), while Step 2 evaluated the objective response rate (ORR) by independent radiological central review (IRCR). Secondary endpoints included progression-free survival (PFS), overall survival (OS), best overall response, disease control rate, clinical benefit rate, safety, and quality of life. Eligible patients had histologically confirmed unresectable ATC, measurable disease per RECIST v1.1, adequate organ function, an ECOG performance status of 0–1, and a life expectancy of over 90 days. Lenvatinib was administered orally at 24 mg once daily, and nivolumab intravenously at 240 mg every two weeks, continued until disease progression or unacceptable toxicity. The Step 2 sample size of 48 patients was determined based on a historical ORR of 23.5% and an expected ORR of 44%, providing 80% power with a two-sided α of 0.05, accounting for a 20% dropout rate. ORR was evaluated in the full analysis set of patients with ATC confirmed by independent pathological central review (IPCR). Results: Between December 2019 and February 2024, 51 patients were enrolled across 10 sites in Japan (Step 1: n=3; Step 2: n=48). No DLTs were observed in Step 1. In Step 2, the median age was 69.5 years (range, 43-89); 46 patients had metastatic disease, 40 had prior history of surgery, and 42 were pathologically confirmed as ATC by IPCR. The ORR by IRCR was 47.6% (95% CI: 33.4-62.3), meeting the primary endpoint. At a median follow-up of 11.6 months, the median duration of response was 12.9 months (95% CI: 6.28 -27.6), median PFS was 5.6 months (95% CI: 5.5-9.1), and median OS was 14.7 months (95% CI: 10.1-29.2). The 1-year OS rate was 56.9%. In Step 2, grade 3 or 4 treatment-related adverse events (TRAEs) occurred in 38/48 patients (79.2%), serious TRAEs in 20 (41.7%), and there was one treatment-related death (2.1%). Conclusions: Nivolumab plus lenvatinib demonstrated a clinically meaningful ORR, meeting the primary endpoint. The combination also achieved favorable PFS and OS with manageable toxicity under appropriate supportive care, suggesting that this regimen may represent a new treatment option for patients with unresectable ATC. Clinical trial information: NCT05696548 .

Differential gene detection and turnaround time between tissue and Guardant360 liquid biopsy next-generation sequencing across solid malignancies.

e15066 Background: Circulating tumor DNA (ctDNA) next-generation sequencing (NGS) is increasingly used for genomic profiling in metastatic solid tumors, yet real-world differences in mutation detection, cross-platform concordance and turnaround time (TAT) across tumor types remain incompletely defined. We compared these metrics between tissue NGS and Guardant360 liquid biopsy in a large, heterogeneous clinical cohort. Methods: We retrospectively analyzed patients with metastatic solid malignancies who underwent tissue NGS and/or Guardant360 testing during routine care at the Cancer Center of Kansas. Cancer type, number of pathogenic or likely pathogenic alterations, and TAT (days from specimen collection to finalized report) were abstracted. Outcomes were summarized overall and by major tumor groups (breast, lung, prostate, colorectal, other). Among patients tested with both types of biopsies, concordance of recurrent mutations was evaluated descriptively. Analyses were conducted under prespecified null hypotheses of no difference in mutation yield or TAT between platforms. Results: A total of 880 evaluable NGS assays were included (227 tissue-based, 653 Guardant360). Guardant360 demonstrated a higher mean mutation yield than tissue NGS overall (mean 6 vs 5; median 5 vs 4). Higher Guardant360 yields were observed in breast (mean 8 vs 4), colorectal (9 vs 5), and other solid tumors (6 vs 5), while yields were similar in lung cancer (6 vs 6) and modestly higher in prostate cancer (5 vs 3). Guardant360 was associated with substantially shorter TAT overall (median 7 vs 20 days; mean 9 vs 67 days), with consistent differences across tumor subgroups; prostate cancer showed the smallest differential (median 7 vs 5 days). Among 136 paired tests, concordance varied by gene and tumor type. TP53, KRAS, APC, EGFR, and PIK3CA were the most frequently shared alterations, though absolute overlap was limited (e.g., TP53 detected on both platforms in 81 cases vs 101 tissue-only and 421 Guardant360-only detections). Overlap patterns reflected tumor biology, with lung cancers contributing most TP53 and EGFR concordance and colorectal cancers predominating in APC and KRAS overlap. Conclusions: In this large real-world, multi-tumor analysis, Guardant360 ctDNA testing demonstrated comparable or higher mutation detection and markedly faster turnaround time than tissue NGS, while revealing biologically meaningful but incomplete genomic concordance. Early liquid biopsy may substantially shorten time to molecular results without compromising detection across diverse solid malignancies, informing treatment selection when tissue testing is delayed or limited. Integrated tissue–liquid sequencing may optimize comprehensive genomic profiling and precision oncology decision-making in metastatic disease.

Diagnostic and prognostic value of plasma D-dimer levels for metastasis and survival in gastric cancer: A systematic review and meta-analysis.

e16089 Background: Gastric cancer (GC) is frequently associated with hypercoagulability, yet the clinical utility of D-dimer as a biomarker remains inconsistent. While elevated D-dimer is linked to venous thromboembolism, its potential to predict distant metastasis and long-term survival in GC is debated. We performed a systematic review and meta-analysis to evaluate the diagnostic accuracy of pretreatment D-dimer for detecting metastasis and its prognostic value for overall survival (OS). Methods: We searched PubMed, Embase, and Cochrane Library databases through January 2026 for observational studies evaluating D-dimer in histologically confirmed GC. Inclusion criteria were studies reporting diagnostic performance (sensitivity/specificity) for metastasis or hazard ratios (HR) for survival. Quality was assessed using QUADAS-2 and NOS. A bivariate random-effects model was used to pool diagnostic metrics. Due to heterogeneous reporting of survival outcomes, prognostic data were synthesized qualitatively. Results: Eight studies comprising 6,189 patients were included. Four studies (n = 5,008) provided extractable data for diagnostic meta-analysis of metastasis (hematogenous, lymph node, or bone). The pooled sensitivity and specificity of elevated D-dimer for detecting metastasis were 0.67 (95% CI, 0.66-0.69) and 0.78 (95% CI, 0.77-0.79), respectively. The diagnostic odds ratio (DOR) was 7.40, indicating patients with metastasis had 7-fold higher odds of elevated D-dimer. The area under the summary receiver operating characteristic curve (SROC AUC) was 0.81, indicating moderate-to-high diagnostic accuracy. Regarding prognosis, 5 studies assessed survival outcomes. Elevated pretreatment D-dimer was universally associated with shorter OS in univariate analyses (P &lt; 0.001). However, multivariate results were mixed: two studies confirmed D-dimer as an independent prognostic factor for 1-year mortality and OS, whereas two others found it dependent on tumor stage. D-dimer levels consistently correlated with tumor load, TNM stage, and vascular invasion. Conclusions: Pretreatment plasma D-dimer demonstrates high specificity and moderate sensitivity for detecting metastasis in gastric cancer, making it a viable, cost-effective initial screening tool. While consistently associated with poor survival, its independent prognostic value varies by study population. Routine D-dimer assessment may aid in risk stratification for occult metastasis. Diagnostic performance of d-dimer for gastric cancer metastasis. Study N Cut-off (mg/L) Sensitivity (%) Specificity (%) AUC (95% CI) Zhang X et al. 2022 3,447 0.91 68.2 76.3 0.77 (0.76–0.79) Wang et al. 2023 454 1.03 69.0 72.8 0.76 (0.71–0.80) Yan et al. 2025 65 1.63 87.1 94.1 0.91 (0.83–0.99) Diao et al. 2014 1,042 1.50 61.9 86.6 0.80 (0.76–0.83)

Comprehensive profiling of antibody-drug conjugate target expression in epithelial ovarian cancer with pre- and post-ADC longitudinal assessment.

5594 Background: Antibody–drug conjugates (ADCs) are increasingly used in ovarian cancer, yet target expression may evolve over time and under treatment pressure. We characterized key ADC targets across disease timepoints and assessed paired changes after ADC exposure. Methods: We retrospectively analyzed epithelial ovarian cancer treated at Yonsei Cancer Center (n=158). Serial specimens obtained at diagnosis, interval debulking surgery, and relapse underwent immunohistochemistry for TROP2, HER2, FOLR1, CLDN6, B7H4, B7H3, and CDH6. Targets were scored 0/1+/2+/3+; overexpression was defined as 2+ or 3+. Longitudinal transitions were summarized using Sankey plots, including pre-/post-ADC comparisons. Results: We analyzed 158 epithelial ovarian cancer patients (HGSC n=120; non-HGSC n=38). In HGSC, the most frequent overexpressed targets (2+/3+) were TROP2 (100%) and FOLR1 (91.5%) at diagnosis, remaining high at relapse (TROP2 93.2%, FOLR1 82.2%), while HER2 overexpression was uncommon (20.3% at diagnosis; 6.9% at relapse). In non-HGSC subtypes, TROP2 was most frequently overexpressed in endometrioid (3+ in 100%) and clear cell tumors (2+/3+ in 100%), whereas mucinous tumors showed predominant HER2 overexpression (2+/3+ in 70%). In ADC-exposed paired pre-/post-ADC specimens, expression of the cognate ADC target generally decreased after exposure, including reduced FOLR1 after MIRV/MORAb-202, reduced HER2 after T-DXd, and reduced CDH6 after R-DXd. Conclusions: TROP2 and FOLR1 are broadly and consistently expressed across timepoints. Non-HGSC exhibits distinct histology-specific patterns, with particularly high HER2 expression in mucinous tumors. ADC treatment resulted in a reduced expression profile of its cognate target protein, which may contribute to the development of ADC resistance; however, further investigation is required to elucidate additional underlying mechanisms.

Prognostic significance of ctDNA detection at the time of relapse during melanoma surveillance.

e21543 Background: Circulating tumor DNA (ctDNA) is an emerging biomarker of progression and relapse in multiple cancers. Our earlier single-center data demonstrated that site of melanoma relapse impacted the sensitivity of ctDNA detection. In this multi-institutional study, we further examined the prognostic implications between ctDNA detection and melanoma relapse site. Methods: A retrospective cohort analysis was performed across three centers using a personalized, tumor-informed ctDNA assay (Natera) on prospectively collected plasma from patients with melanoma from December 2021 to January 2025 with longitudinal follow up. Inclusion criteria were patients with no evidence of disease (NED) following definitive surgery or clinical remission with systemic therapy; and undetectable ctDNA. Sensitivity of ctDNA detection at biopsy- and/or radiographically confirmed relapse was evaluated, along with overall survival (OS) from date of relapse to date of death or last follow-up. Analyses were stratified based on anatomic sites of relapse. Logistic regression was used to evaluate predictors of ctDNA detection. Log-rank p-values and Cox proportional hazards were calculated for OS measurements. Results: Of 284 patients on longitudinal MRD surveillance, 78 (27.5%) had confirmed relapse. Of these, 75 (96.2%) had NED following surgery and 3 (3.8%) had clinical remission with systemic therapy, with 49 (62.8%) resected stage II/III, 26 (33.3%) resected stage IV, and 3 (3.8%) unresectable stage III/IV in remission. Primary sites included 61 (78.2%) cutaneous, 12 (15.4%) mucosal, and 5 (6.4%) unknown. ctDNA was detected at relapse in 40/78 patients (51.3% sensitivity). Similar to prior data, sensitivity varied by relapse site with 84.6% (22/26) for lymph node (LN) relapse and 28.6% (6/21) for skin/soft tissue (SST) relapse. LN metastases (OR = 10.4, 95% CI 3.1 – 34.8; p &lt; 0.001) and multiple (2+) metastatic sites (OR = 4.43, 95% CI: 1.13 – 17.4; p = 0.037) were associated with detectable ctDNA at relapse, while SST metastases (OR 0.27 (95% CI: 0.09 – 0.80; p = 0.021) were associated with an undetectable ctDNA. Median follow-up time from relapse was 13.4 months. Survival analysis from time of relapse showed decreased OS in those with undetectable versus detectable ctDNA (log-rank p = 0.041; HR 0.321, 95% CI: 0.102 – 1.011), including in the SST relapse subgroup (log-rank p = 0.0016; HR not estimable). Conclusions: Detectable ctDNA at the time of melanoma relapse appears to be associated with inferior OS. ctDNA sensitivity varied by site of relapse, with improved survival observed for those with undetectable ctDNA at relapse. These findings support the prognostic relevance of ctDNA while highlighting important site-specific limitations, warranting validation in larger prospective cohorts.

Protocatechuic acid prevents obesity caused by long-chain saturated fatty acid-induced inflammation in mouse microglia via inhibition of the NF-κB pathway

Long-chain saturated fatty acids (LCSFAs), abundant in animal fats, can directly activate microglia and elicit inflammatory responses. Excessive intake of LCSFA-rich high-fat diets (HFDs) has been linked to microglial activation in the brain (particularly within the hypothalamus, a central regulator of energy metabolism) and to metabolic disorders, including obesity. Here, we report that protocatechuic acid (PCA) suppressed the inflammatory response in murine microglia induced by LCSFAs. PCA inhibited the ubiquitin-proteasome degradation of IκBα induced by LCSFAs, suppressing the nuclear translocation of NF-κB and the expression of pro-inflammatory cytokine genes, which was indicated to be attributed to the suppression of I kappa B kinase. In addition, PCA prevented obesity by inhibiting the accumulation of activated microglia in the ARC of HFD-fed mice. This study is the first to demonstrate that PCA suppresses the inflammatory response of microglia induced by LCSFAs and ARC inflammation in HFD-fed mice. These findings provide new evidence and insights into the mechanisms by which polyphenols, including PCA and its analogs, ameliorate diet-induced obesity.

Gate-width scaling modulates electron velocity in AlGaN/GaN HEMTs

In this Letter, the impact of gate-width (Wg) scaling on the electron transport and electrical characteristics of GaN HEMTs is investigated using an experiment-validated coupled drift-diffusion and Monte Carlo framework. For gate length (Lg) = 0.2 μm devices with Wg scaled from 200 to 80 μm, the peak ve increases by 10%, leading to higher normalized transconductance (gm) and Id, primarily due to reduced lattice temperature (TL) and the polarization Coulomb field scattering. Meanwhile, the maximum cutoff frequency (fT,max) slightly degrades because the reduced total transconductance (gm,tot) increases the extrinsic delay (τext). These results provide a quantitative transport-based explanation of the Wg-dependent performance in GaN HEMTs.

Surface-engineered 3D printed scaffolds with tailored nano topography: In vitro degradation and in vivo studies

High‐Performance Flexible Pyroelectric Energy Harvesting System Enabled by Light‐Driven Thermomechanical Coupling in Liquid Crystal Elastomer (Adv. Mater. 35/2026)

Mapping peak flowering phenology of rapeseed (Brassica napus) in North Dakota using Landsat 8 and Sentinel-2

Abstract Detecting the peak flowering time of rapeseed ( Brassica napus ) is essential for optimizing crop management, improving yield forecasts, and enhancing pollination services. This study aimed to map and analyze the spatial and temporal patterns of peak flowering in spring rapeseed across North Dakota using satellite remote sensing. We utilized multi-year optical imagery from Landsat 8 (2013–2024) and Sentinel-2 (2018–2024), applying the Normalized Difference Yellow Index (NDYI)—an index sensitive to yellow floral reflectance—to identify peak flowering dates. Our results indicate that peak flowering dates generally cluster around Julian day 200 (mid-July). Sentinel-2 provided more temporally consistent and spatially detailed estimates than Landsat, owing to its higher revisit frequency (5–7 days versus 16 days) and finer spatial resolution (10 m versus 30 m). On average, Sentinel-2 detected slightly later flowering dates (Julian days 203–230) compared to Landsat (196–213). Median peak flowering dates from Landsat ranged between Julian days 191 and 221, whereas Sentinel-2 medians were narrower and more closely aligned with the mid-July window, spanning Julian days 202 to 216. These differences underscore Sentinel-2’s superior capacity to capture flowering events precisely, while Landsat’s coarser temporal resolution occasionally resulted in earlier or delayed detections. Overall, the NDYI-based approach effectively captured flowering dynamics, revealing both inter-annual variability and field-level heterogeneity. These findings demonstrate the value of NDYI remote sensing for monitoring rapeseed phenology and suggest its broader applicability in agricultural and ecological contexts.

Structural and mechanistic basis of ubiquitous bacterial kinase signaling identifies PorX as a noncanonical substrate in Porphyromonas gingivalis